Seasonal variations in ice deformation and basal motion across the tongue of Haut Glacier d'Arolla,...

Seasonal variations in ice deformation and basal motionacross the tongue of Haut Glacier drsquoArolla Switzerland

Ian WILLIS1 Douglas MAIR2 Bryn HUBBARD3 Peter NIENOW4 Urs H FISCHER5

Alun HUBBARD6

1Scott Polar Research Institute and Department of Geography University of Cambridge Cambridge CB2 1ER EnglandE-mail iw102cuscamacuk

2Department of Geography and Environment University of Aberdeen Elphinstone Road Aberdeen AB24 3UF Scotland3Centre for Glaciology Institute of Geography and Earth Sciences University ofWales Aberystwyth SY23 3DBWales

4Department of Geography andTopographic Sciences University of Glasgow Glasgow G12 8QQ Scotland5Laboratory of Hydraulics Hydrology and Glaciology ETH-Zentrum CH-8092 Zulaquo rich Switzerland

6Department of Geography University of Edinburgh Drummond Street Edinburgh EH8 9XP Scotland

ABSTRACT Records of surface motion englacial tilt and repeat inclinometry are usedto determine patterns of surface internal and basal motion across the tongue of HautGlacier drsquoArolla Switzerland over temporal scales ranging from days to months Findingsare interpreted with reference to contemporaneous measurements of subglacial water pres-sures and prior knowledge of the glacierrsquos subglacial drainage-system structure Long-terminclinometry results show pronounced extrusion flow over a subglacial drainage axis withbasal velocities up to twice those measured at the glacier surface Deformation profiles aremore conventional away from the drainage axis with basal velocities sup160^70 of surfacevelocities Comparison of long-term tilt rates from repeat inclinometry and englacial tilt-meters shows close correspondence Englacial tiltmeter data are used to reconstruct internalvelocity profiles and to split surface velocities into internal deformation and basal motioncontributions over spring summer and autumnwinter periods Although spatial patternsof surface movement are similar between periods patterns of internal and basal motion arenot Results are interpreted in terms of the location of sticky and slippery spots with tem-porally changing patterns of basal drag reflecting changing distributions of water pressure

INTRODUCTION

Glaciers move by a combination of internal deformation andbasal motion the latter involving sliding at the ice^bed inter-face andor the deformation of subglacial sediment Know-ledge of the relative importance of these mechanisms howthey vary spatially and temporally and what controls thisvariability is crucial to an understanding of glacier and ice-sheet dynamics and the representation of these mechanismsin numerical models Several studies have measured surfacemotion calculated internal deformation from Glenrsquos flow lawand estimated basal motion as the residual (see Copland andothers1997b Gudmundsson and others1999 and referencestherein) The problem with this method is that the accuracyof Glenrsquos flow law in these circumstances is unknown since itassumes that the local shear strain rate is controlled by thelocal shear stress and it ignores variations in ice rheologyand the effects of longitudinal stress gradients (Blatter andothers 1998) Other studies have measured surface motionand then used inverse force-balance techniques to calculatebasal drag as a surrogate for basal motion (Hooke andothers 1989Van derVeen and Whillans1993 Iken and Truf-fer1997 Mair and others 2001) Although longitudinal stressgradients are accounted for in this approach they are cal-culated on the basis of surface strain rates that are assumedto be constant with depth Consequently the method canonly be applied over spatial scales greater than a few ice

depths Other studies have attempted to measure basalmotion directly in natural cavities artificial tunnels or atthe base of boreholes using photography video or purpose-built instruments (see Gudmundsson and others 1999 andreferences therein Hubbard 2002) Such studies tend tomeasure basal motion over a spatially restricted area andover time-scales of hours to weeks and the results cannoteasily be extrapolated to broader spatial scales or longertime-scales Furthermore such studies have not usually beencombined with measurements of surface motion and havetherefore not been able to determine the contributions ofinternal deformation and basal motion to overall flow

Repeat borehole inclinometry has successfully been usedto examine the contribution of internal deformation (andby subtraction basal motion) to net glacier motion overtime-scales of months to years (see Copland and others1997b and references therein Harbor and others 1997Harper and others1998) However on slow-moving glacierssuch as Haut Glacier drsquoArolla Switzerland repeat inclinom-etry cannot be used to resolve the components of surfacemotion at time-scales less than about 1year (Copland andothers 1997b) Recently englacial tiltmeter records col-lected by data logger at high temporal resolution have beenused to reconstruct the ice-deformation profile and the con-tributions of ice deformation and basal motion to surfaceflow (Gudmundsson and others 1999) While the techniquewas applied to just one borehole on the centre line of Unter-

Annals of Glaciology 36 2003 International Glaciological Society

157

aargletscher Switzerland over a 2 year period the highsensitivity of tiltmeters and their ease of manufacture meansthat the technique has the potential to work over large areasand over much shorter time intervals

The overall aim of this paper is to use surface motionmeasurements englacial tiltmeter records and repeat inclin-ometry data to determine patterns of surface internal andbasal motion across the tongue of a temperate valley glacierover a variety of temporal scales ranging from days tomonths We interpret the findings with reference to contem-poraneous measurements of subglacial water pressures andin terms of what we already know about the hydrology ofthe glacier

METHODS

Field site

Haut Glacier drsquoArolla is a sup163 km2 sup140 km long temper-ate valley glacier in Valais Switzerland (45sup358rsquo N 7sup332rsquo E)ranging in elevation from sup12560 to 3500 m asl (Fig 1)During each of the summers of 1992^99 an array of bore-holes was drilled sup115 km from the glacier snout and east ofthe glacier centre line (Fig 1) On average sup125 boreholeswere drilled each year ranging in depth from sup130 m nearthe margin to 4130 m on the centre line and these havebeen used to study the englacial and subglacial hydrology(B Hubbard and others 1995 1998 Copland and others1997a Kulessa and Hubbard1997 Gordon and others19982001) and geochemistry (Lamb and others1995Tranter andothers 2002) of this part of the glacier Boreholes drilled in1995 and 1996 were also used to calculate annual patternsof surface internal and basal motion in a cross-section of

the glacier from the centre line to the eastern margin usingrepeat inclinometry techniques (Harbor and others 1997)Analysis of several boreholes drilled in this area in 1998and1999 forms the basis of the work presented in this paperThe boreholes were drilled with reference to a subglacialdrainage axis that ran down the eastern half of the glacierduring the early to mid-1990s (Sharp and others 1993Hubbard and others 1995) The axis location is controlledlargely by the glacier geometry but also by the position ofmoulins Changes in glacier geometry and moulin positionmay have altered the precise location of this axis since it wasfirst identified but for the purposes of this paper we refer tothe axis as it existed in the early to mid-1990sThe axis trans-ports surface-derived waters andevolves from a distributed toa channelized morphologyduring the early summer (Gordonand others1998 Nienow and others1998) In late summer itconsists of a channel bounded laterally by subglacial sedi-ments During the day water and water-pressure waves areforced out of the channel into the surrounding sedimentsand possibly a thin film between the glacier sole and thesediments and at night water and water-pressure waves aredrivenback to the channel (Hubbard and others1995) Pres-sure fluctuations in the channel affect areas of the bed up tosup170 m either side (Hubbard and others1995)

Inclinometry

In late May and earlyJune1998 four boreholes were drilledto the bed at seven sites on the glacier tongue giving 28boreholes in total (Fig 1) Sites UA MA and LA werelocated along the centre of the subglacial drainage axisSites UB MB and LB were positioned sup170 m west of theaxis Site MC was located a further sup170 m to the west near

Fig 1 Haut Glacier drsquoArolla showing the location of the seven borehole sitesThe sites are labelled according to their position alongthe glacier (U upper M middle L lower) and across glacier (A over the axis C near the centre line B between the axis andthe centre line) Contours are shown for the glacier surface (solid) and bedrock (dashed)

Willis and others Basal motion across tongue of Haut Glacier drsquoArolla

158

the glacier centre line All boreholes were inclinometeredwithin 2 days of being drilled using a MountainWatch Incdigital borehole inclinometerThe inclinometer consists of aflux-gate magnetometer and two tilt transducers housed in a2 m long stainless-steel tube The inclinometer is connectedvia a control box to a portable computer The tilt and azi-muth of the inclinometer were recorded at intervals of 1malong the boreholeThe measurements were made as the in-strument was both lowered down and raised up the bore-hole and the final profile was constructed from the meanof the two sets of readings This gives measurements thatare accurate to sup105 of ice depth (cf Copland and others1997b) Software developed by Icefield Instruments Incwas used to resolve the tilt and azimuth data into a north^south profile (approximately along glacier) and an east^west profile (approximately across glacier) and trigonom-etry was used to produce a profile in the flow direction

In August 1999 six of the original 28 boreholes (oneborehole each at six of the seven sites) were re-drilled tothe bed using a cable-following drill tip Unfortunatelynone of the boreholes at site UA could be re-drilled to thebed because the water in the boreholes drained when thedrill tip reached an englacial connection at sup180 m depthmaking further drilling impossible The six boreholes werere-inclinometered within 2 days of being re-drilled usingthe same procedure as in 1998 (Table1)The data were againused to calculate along-glacier across-glacier and flow-dir-ection profiles Subtraction of the original profiles from thenew profiles was used to determine the vertical variations invelocity through the glacier

Borehole instrumentation

In MayJune 1998 a suite of instruments was deployed ateach of the seven sites including a pressure transducer (tomonitor subglacial water pressure) and an englacial tilt-meter at 50 and 90 of ice depth (with which to con-struct the internal deformation profile) In August 1998additional englacial tiltmeters were placed at 75 of icedepth at site MB and 100 of ice depth (05 m above thebed) at sites MA and MB in order to reconstruct theinternal deformation profiles at these sites more accuratelyEach pressure transducer consists of a GEMS Sensors2000A 1^16 bar transducer in stainless-steel housing Theywere calibrated in the field by lowering and raising them inan inclinometered water-filled borehole Repeat calibra-tions suggest that they have an accuracy of sect025 m Eachenglacial tiltmeter consists of a single Fredericks Companyproportional non-linear dual-axis miniature electrolytic tilt

sensor (0717 2201) housed and sealed into a 08 m long stain-less-steel rod The sensors were similar to those used tomonitor subglacial sediment deformation and described byBlake and others (1992) and Porter and Murray (2001) Theoutput from each tiltmeter consists of two voltage signals(one for each axis) These are converted to tilt and relativeazimuth data using a purpose-written inversion program(based on Blake and others1992)

Surface motion

The tops of all 28 boreholes were surveyed within 4 days ofbeing inclinometered in MayJune 1998 and were resur-veyed within 4 days of being re-inclinometered in August1999 (Table 1) These data were used to calculate surfacevelocities over the whole measurement period Additionalsurveys were made at regular intervals throughout thespring and summer of 1998 and 1999 from which seasonalsurface velocities were derived Surveys were conductedusing a Geodimeter 400 total station from a survey stationat the eastern margin of the glacier tongue with reference tothree stationary reference points fixed within the Swiss Grid(Fig 1) The instrument accuracy is sect (2 mm +3 ppm)which corresponds to 5sect3 mm over the distances sur-veyedThis corresponds to velocity errors of lt5 over theshortest 10 day time period considered here and negligibleerrors over longer intervals

RESULTS AND DISCUSSION

Repeat inclinometry

The vertical variations in horizontal velocity (boreholedeformation profiles) at the six sites are shown in Figure 2The boreholes at sites MB MC LB and UB which arelocated away from the drainage axis show classic deform-ation profiles with greatest velocities at the surface decliningvery gradually and quasi-linearly with depth to sup170 of icedepth and then more steeply to the bed The ratio of basalvelocity to surface velocity was sup150 at site UB sup160 atsites MB and MC and sup165 at site LB These values aresimilar to those found for this part of the glacier by repeatinclinometry between 1995 and 1996 (Harbor and others1997) Similar results were also found on WorthingtonGlacier Alaska USA where annual basal velocities ac-counted for 60^70 of the annual surface motion in an areaaround the centre line of the upper ablation area where icedepths were sup1200m (Harper andothers1998) Slightlyhigherratios were found in the ablation areas of both AthabascaGlacier Canada (ice depths sup1300m) and Storglacialaquo renSweden (ice depths sup1150 m) of sup185 on the centre line andsup170^75 halfway to the margin (Raymond1971 Hooke andothers1992)

The profiles at sites MA and LA located on the drainageaxis contrast sharply with expected profiles (Fig 2) At siteMA velocities were fairly constant from the surface tosup150 ice depth and then increased to sup185 ice depthThereafter velocities decreased sharply and linearly overthe remaining 13 m to the bed At site LA velocitiesincreased to sup180 ice depth andthen also decreased sharplyand linearly over the remaining 19 m to the bed Basalvelocity is similar to surface velocity at site LA but twotimes greater than surface velocity at site MA Thus theboreholes at sites MA and LA are characterized by local

Table 1 Inclinometry and surface survey dates Dates aregiven as days since 1 January 1998

Site Depth Inclinometry Re-inclinometry Survey Resurveydate date date date

m

MA 1029 Day139 Day 598 Day139 Day 595MB 1247 Day139 Day 591 Day142 Day 595MC 1300 Day143 Day 591 Day147 Day 595LA 960 Day147 Day 592 Day147 Day 595LB 1139 Day147 Day 592 Day147 Day 596UB 1339 Day157 Day 595 Day156 Day 596

159


extrusion flow Local extrusion flow was also recorded onthis glacier in the vicinity of site MA between 1995 and1996 (Harbor and others1997) In this earlier study surfacespeeds were sup1825 m a^1 but maximum speeds of sup1975m a^1 were reached at sup150 of ice depth (Harbor andothers1997 fig 2D) In this study surface speeds were com-parable but maximum speeds of sup116 m a^1 were reached atsup185 of ice depth

Extrusion flow was also implied from long-term englacialtiltmeter measurements in a sup1300 m long borehole drilled onthe centre line in the ablation area of Unteraargletscherwhere annual basal motion was estimated to be sup1120 ofannual surface movement (Gudmundsson and others 1999)Similarly extrusion flow was measured in a126 m deep bore-hole down-glacier from a riegel on Storglacialaquo ren where basalmotion was sup1500 of surface motion for a 3 week period inJuly (Hooke and others1987)

We interpret our observed local extrusion flow in terms ofenhanced basal motion associated with water flow along thepreferential drainage axis Penetrometer tests at the base ofboreholes drilled abovethe axis suggest that a subglacial sedi-ment layer is either very thin or absent so basal motion likelycomprises sliding rather than sediment deformation Further-more we know from visits in November January and March

that water continues to flow from the glacier throughout thewinter and chemical analysis of the water suggests it flowssubglacially Thus we envisage the continuous flow of wateralong the drainage axis throughout the year is sufficient tolubricate the bed here and promote rapid sliding Numericalmodelling experiments for this section of the glacier showthat previously observed annual patterns of local extrusionflow can be recreated assuming a composite time-weightedaverage of 20 weeks no sliding 20 weeks moderate slidingand 12 weeks enhanced sliding (A Hubbard and others1998) Local extrusion flow has also been modelled for anidealized sinusoidal bed and may be expected to occur bothabove the crest and above the trough of a sinusoid (Gud-mundsson1997) Our observed profiles show some similaritywith the modelled patterns above the crest of a sinusoid witha velocity maximum just above the bed (Gudmundsson1997fig 2) Further modelling is required to examine whether ourobserved local extrusion flow is due to the occurrence of aslippery spot a bump or some combination of the two

Englacial tiltmeters

The dual-axis tilt and azimuth data were used to resolve thetilt in the direction of ice flow using the procedure described

Fig 2 Horizontal velocity profiles in the ice-flow direction of six boreholes between MayJune1998 and August1999 Horizontalexaggeration 610


160

Fig 3 Long-term records of englacial tilt in the ice-flow direction at six borehole sites thick dark lines 50 ice depth thin darklines 90 ice depth thick light lines 100 ice depth thin light lines 75 ice depth A vertical tiltmeter has a tilt of zeronegative tilts mean the meter is dipping down-glacier positive tilt occurs when the meter is dipping up-glacier

161


by Blake and others (1992) and Mair and others (personalcommunication 2002) The long-term records of englacialtilt in the ice-flow direction at the six sites are shown in Fig-ure 3 Despite gaps in these records the data indicate that atall sites with the possible exception of site MC the long-term tilt rate is greater at 90 ice depth than at 50 icedepth It is difficult to identify the long-term tilt rate at50 ice depth at site MC due to the erratic behaviour

during spring and summer 1999 At sites MA and MB thelong-term tilt rates at 100 ice depth are greater than at90 And at site MB tilt rates at 75 ice depth are similarto those at 50The records can be split into distinct periodsduring which certain trends can be observed The recordsgenerally show long periods of steady linear tilt rates duringthe autumnwinter period (August 1998^March 1999) Thetilt records during both the 1998 and 1999 springsummer

Fig 4 Long-term records of subglacial water pressure at seven borehole sites iob on vertical axes refers to ice overburden pressureThe horizontal line at 100 iob is therefore the flotation level


162

periods however are more erratic They show (i) medium-term (days to weeks) trends in tilt rate (ii) short-term (hoursto days) fluctuations in tilt rates superimposed onthe medium-term trends and (iii) short-term jumps in tilt rate that breakthe medium-term trends This erratic behaviour occurs atall recorded depths in both springsummer periodsalthough it is particularly marked at 50 ice depth at siteMC during the 1999 springsummer We interpret the tilt-meter records in terms of slow ductile deformation that pre-dominates during the autumn and winter and ductiledeformation interspersed with brittle fracture associatedwith ice-quake activity during the spring and summerBrittle fracture was particularly marked during springevents and is discussed in more detail by Mair and others(personal communication 2002) It is possible that some ofthe erratic behaviour during spring and early summer 1998may have been due to movement of the tiltmeter within theborehole rather than movement of the borehole itself How-ever our experience suggests that boreholes tend to closearound instruments and cables within a few weeks unless theyare kept open by reaming or manual probing The erraticbehaviour during late summer 1998 and spring and summer1999 must therefore be due to movement of the borehole notthe instrument We also rule out the possibility of electronicnoise to explain the erratic behaviour since the short-termfluctuations and jumps affected each instrument uniquelyThere were instances when one tiltmeter was affected butanother in the same borehole was unaffected even thoughthe same data logger controlled the two instruments Con-versely there were occasions when several instrumentsresponded in similar ways even though they were in separateboreholes controlled by different data loggersThere are fourinstances where tilt rates seem unusually large These are at90 ice depth at site MB in late summer 1998 (sup13sup3 d^1) andspring1999 (sup1025sup3 d^1) and at both 90 ice depth (until 5September) and 50 ice depth (until 22 September) at siteMC in early autumn 1998 (both sup1025sup3d^1) We are unsureof the reasons for these unusually large tilt rates anddiscount these periods from our calculations of seasonaldeformation profiles presented below

Subglacial water pressures

Long-term records of subglacial water pressures are shownin Figure 4 Despite some gaps records show generally highand stable water pressures around ice overburden duringthe autumnwinter Records at sites MA and LA on thedrainage axis are very stable over the autumnwinter Pres-sures at sites away from the axis are slightly more variablePressures at sites MC and UB away from the axis graduallyrise during autumn and early winter1998 reaching a peak onday 340 Pressures then drop dramatically at site MC andmore gradually at site UB until day 343 before rising rapidlyagain to ice overburdenby day 350 at site MC while continu-ing to fall though at a slower rate at site UB The record atsite UB shows a further disturbance between days 375 and390 during which pressures rise to ice overburden beforedeclining once more

During the spring and summer records are generallymore variable than during the autumnwinter The rapidwater-pressure fluctuations that are particularly marked atsites MA on the axis and LB 70 m west of the axis duringspring1998 centred on days158 and173 are associated withspring speed-up events Similarly the fluctuations in spring

1999 beginning around day 552 are also associated with aspring event These spring events are discussed in moredetail by Mair and others (in press)

We interpret many of the short-term (daily) water-pres-sure fluctuations especially during the spring and summerin terms of melt- and rain-induced variations in surfacewater inputs which are the cause of glacier motion events(Mair and others in press) However other pressure fluc-tuations including those during the autumnwinter butothers during the summer (eg the rapid drop at site MCand the fast rise at site UA both centred on day 560 andthe sudden drop at site LB centred on day 570) are causedby boreholes connecting to and disconnecting from a sub-glacial drainage system and may be an effect of glaciermotion events

Comparison of inclinometer and tiltmeter data

In this subsection we compare the long-term tilt rates meas-ured by the tiltmeters with those determined by repeat inclin-ometry At each of the six sites the borehole containing thetiltmeters was not the same as that used for repeat inclinom-etry However the boreholes used for comparison werelocated within 2 m of each other at the surface whichshould produce negligible discrepancies between the twosets of measurements The time interval over which tilt rateswere determined using the inclinometry method variedbetween 438 and 459 days but due to the erratic behaviourof some of the tilt cells during the springsummer periodstilt rates were determined from the tiltmeters for theautumnwinter period only For sites MA MB MC andUB records were 151^218 days long For sites LA and LBhowever records were only 33^36days long due to missingdata For these sites where erratic behaviour was negligibleduring the 1999 springsummer we also calculated long-term tilt rates from the tiltmeter data over longer time inter-vals (autumn 1998^autumn 1999) but these were virtuallyidentical (sect3) to those determined for the short autumnwinter period aloneTo be consistent across all sites we usethe autumnwinter data only in our comparison as itappears that these are generally representative of the over-all annual deformation pattern

The procedure for calculating the tilt rates from theinclinometry data is as follows First at the equivalent depthin the inclinometry borehole at which a tiltmeter is locatedthe x y z coordinates of the borehole at the time of initialinclinometry (t ˆ1) and the time of subsequent re-inclinom-etry (t ˆ 2) are used to calculate the flow direction Secondfor both t ˆ 1 and t ˆ 2 the x y z co-ordinates at theappropriate depth and1m above that depth are used to calcu-late the tilt angle of the borehole resolved in the flow direc-tion Third the change in tilt angle in the flow direction iscalculated between the two time periods and expressed inunits of sup3 a^1 Finally these steps are repeated for 3 5 7 and9 m long borehole sections centred on the original 1m longsection This produces five estimates of tilt-angle changewhich are used to calculate a mean

Table 2 shows the data used to derive the two sets of tilt-rate estimates that are compared in Figure 5There is a verygood match between the tilt changes derived from the tilt-meters and those calculated from the inclinometry recordswith an overall coefficient of variation of 094 The meanabsolute difference is just 10sup3 a^1

163


Seasonal surface internal and basal velocities

In this subsection we divide the total time between inclinom-etry measurements (May 1998^August 1999) into five sea-sonal periods on the basis of the surface-velocity englacial-tiltmeter and subglacial water-pressure measurements Therationale behind this is that the hydrological and dynamicregimes of the glacier appear to be similar within these fiveperiods but different between them Previous modelling ofthe glacier flow regime at Haut Glacier drsquoArolla suggests thisis a reasonable assumption (A Hubbard and others 1998)The five periods are

Spring events 1998 4^7 and 21^28 June (days 155^158and172^179) ˆ 10 days

Summer 1998 30 May^14 August (days 150^226) ˆ66 days

Autumn1998winter199914 August^20 June (days 226^536) ˆ 310 days

Spring event 1999 29 June^9 July (days 545^555) ˆ10 days

Summer 1999 20 June^16 August (days 536^593) ˆ47 days

In the rest of this subsection we use the surface-velocityand englacial-tiltmeter data to determine patterns of sur-face internal and basal motion for the five seasonal periodsFor each period surface velocities were determined fromthe survey data internal velocity profiles were determinedfrom the englacial-tiltmeter data as explained below andbasal velocities were calculated as the residual Internalvelocity profiles were determined for each site as follows

First the tiltmeter records over the autumnwinter periodwere used to construct the entire velocity profile for the yearRegression lines were fitted to the tilt-angle vs time data andthe slopes of the lines were used to calculate the long-term tiltrates Then several velocity profiles were constructed underthe assumption that each tiltmeter was representative of cer-tain depth ranges For example a profile was based on theassumption that the tiltmeter at 90 ice depth was represen-tative of the depth range100^90 that the meter at 50 ice

depth was representative of 90^50 and that there was notilt from 50 ice depth to the surface At the other extremea profile was based on the assumption that the tiltmeter at90 ice depth was representative of the depth range 100^70 that the meter at 50 ice depth was representative of70^30 and that there was no tilt from 30 ice depth to thesurface The resulting internal velocity estimates were com-pared with those measured by repeat inclinometry over thewhole measurement period to find the best match Becauseof the anomalous inclinometry profiles at sites MA and LAon the drainage axis (above) we could only construct theprofiles for the remaining sites MB MC LB and UB usingthis methodThe depths over which the tiltmeters were foundto be representative are shown inTable 3

We then assumed that the representative depths showninTable 3 applied to the other seasonal time periods as wellFor each of the seasonal time periods and for each tiltmeterthe tilt value at the beginning of the period was subtractedfrom the tilt value at the end of the period to give totalchange in tilt As mentioned above the four instances wheretilt rates seem unusually large (ie site MB at 90 ice depthin late summer 1998 and spring 1999 and site MC at both90 and 50 ice depth in early autumn 1998) wereexcluded from the analysis Furthermore where a largejump in tilt occurred presumably due to brittle fracturethe raw data were corrected to remove the jump as it wouldbe misleading to extrapolate the effects of a local fractureacross several tens of metres of ice depthThe changes in tiltwere divided by time to yield mean tilt rates Using these tiltrates and the representative depth ranges given in Table 3the internal velocity profiles were calculated for sites MBMC LB and UB for each time period

The resultant surface internal and basal velocities forthe four sites are shown in Figure 6 For each of the five timeperiods discussed above the average water pressures for allseven sites were also calculated (Table 4) To help identify thegeneral patterns of behaviour data for the 1998 and 1999spring events and for the 1998 and 1999 summers were aver-aged to produce the spring and summer data shown in Figure6 andTable 4The data show the following characteristics

Table 2 Comparison of tilt rates determined by tiltmeters andinclinometry Dates are given as days since 1 January 1998

Site Depth Tiltmeter Tilt ratedates Tiltmeter Inclinometer

sup3 a^1 sup3 a^1

MA 50 Days 238^456 95 90 (sect05)MA 90 Days 238^455 114 101 (sect77)MB 50 Days 238^456 64 54 (sect05)MB 90 Days 238^456 56 68 (sect24)MC 50 Days 305^456 15 15 (sect04)MC 90 Days 249^455 48 58 (sect14)LA 50 Days 238^274 30 25 (sect02)LA 90 Days 238^274 131 97 (sect13)LB 50 Days 240^274 11 18 (sect17)LB 90 Days 241^274 155 145 (sect19)UB 50 Days 238^420 10 04 (sect05)UB 90 Days 241^456 75 68 (sect24)

Fig 5 Comparison of tilt-angle changes measured by the tilt-meters with those determined by repeat inclinometry

Table 3 Depths () over which tiltmeters are representative

Site Tilt at 100 Tilt at 90 Tilt at 75 Tilt at 50 No deformation

MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0

Excluding spring events


164

Spring eventsDuring the spring events average water pressures are justbelow or just over ice overburden at the upper and middle sitesMA MB MC UA and UB but well below ice overburden atsites LA and LB located down-glacier (Table 4) Surfacevelocities were everywhere 4006 md^1 about twice as highas the summer average (Fig 6) Basal motion makes up mostof the surface motion at site UB and to a lesser extent sitesMB and MC but only a small component of surface motionat site LB located down-glacier Thus the evidence suggeststhat the spring events are directly forced by high water pres-sures in the upper tongue but indirectly forced by pushingfrom up-glacier in the lower tongue where ice deformation isunusually high due to a large longitudinal stress gradient

SummerDuring the summer surface velocities typically drop to lessthan half their values during the spring events (Fig6) Basalmotion makes up sup150 of surface motion at sites LB andUB but only 15 of surface motion at site MB At sitesMA MB and MC in the middle array average water pres-sures are substantially lower than they are during the springevents especially at sites MA and MC dropping to belowice overburden (Table 4) At sites UA and UB in the upper

array mean water pressures are similar to those recordedduring the spring events although they remain over 90of ice overburden At sites LA and LB in the lower arraywater pressures remain well below ice overburden decreas-ing at site LA but increasing at site LB Thus the lowervelocities during the summer compared to those during thespring events are due to a reduction in basal motion at sitesMB and UB and a decrease in ice deformation at site LBBasal motion drops to sup125 of its spring-event value atthe upper array (site UB) where water pressures remainhigh (sites UA and UB) It drops to sup112 of its spring-eventvalue in the middle array (site MB) where water pressuresdrop slightly at site MB but drop substantially either side(sites MA and MC) Finally in the lower array (site LB)basal motion is the same as during the spring events wherewater pressures remain well below ice overburden (sites LAand LB) Thus the evidence suggests that site MB becomesa sticky spot during the summer as water pressures dropeither side even though local water pressures remain high

AutumnwinterDuring the autumnwinter surface velocities drop to sup175of their summer value (sup135 of their spring-event value)(Fig 6) At site MB the relative contributions of internaldeformation and basal motion to surface movement aresimilar to those during the summer with negligible basalmotion Conversely at sites LB and UB both the absolutemagnitude and the relative contribution of basal motion tosurface movement are greater during the autumnwinterthan during the summer At site LB basal motion is higherthan during the spring events At the middle array waterpressures have risen to greater than ice overburden at sitesMA and MC but have dropped to sup190 of ice overburdenat site MB (Table 4) At site LA in the lower array averagewater pressure increased substantially from its average sum-mer value to near ice overburden In the upper array theaverage water pressure at site UB was much lower thanduring the summer (sup145 of ice overburden) but therewas an event during earlyJanuary during which water pres-

Fig 6 Surface internal and basal velocities during the spring events the summer period and the autumnwinter period for sitesMB MC LB and UBThe spring-events period and the summer period are the average of data for 1998 and1999

Table 4 Mean water pressures ( ice overburden) duringthe seasonal periods Spring and summer data are the averageof 1998 and 1999

Site SeasonSpring Summer Autumnwinter

MA 995 782 1153MB 1097 994 915MC 1061 649 956LA 789 638 962LB 794 892 UnknownUA 1048 1098 UnknownUB 968 933 447

165


sures rose to over ice overburden for 10 days (Fig 4) Thisevent did not occur at the other sites Unfortunately we donot know what average water pressure was at site UA but ifthe behaviour at the other sites on the drainage axis (sitesMA and LA) is representative pressures would also haverisen to around ice overburden Thus the evidence impliesthat site MB remained a sticky spot during the autumnwinter as pressures dropped even though pressures hadrisen at sites MA and MC either side By contrast sites LBand UB were relatively slipperyWe do not know the patternof water pressure at site LB during the autumnwinter butthe pattern at UB suggests that the slippery spot there mayhave been due to a short-term (10 day) increase in waterpressure to overburden rather than simply a sustained highwater pressure

CONCLUSIONS

We have used records of surface motion englacial tilt andrepeat inclinometry to determine patterns of surfaceinternal and basal motion across the tongue of Haut GlacierdrsquoArolla over a variety of temporal scales ranging from daysto months We have interpreted the findings with referenceto contemporaneous measurements of subglacialwaterpres-sures and in terms of prior knowledge of the subglacialdrainage-system structure

The long-term inclinometry data show unusual internaldeformation profiles over a subglacial drainage axis withpronounced extrusion flow This confirms earlier resultsreported for this part of the glacier by Harbor and others(1997) but shows even more pronounced extrusion flow thatextends further down-glacier than previously known Ourdata add to the increasing body of evidence that suggeststhat extrusion flow may be a common feature of many gla-ciers where local conditions promote marked reductions inbasal drag (Hooke and others 1987 Gudmundsson andothers 1999) Away from the subglacial drainage axisdeformation profiles are more conventional with basalvelocities sup160^70 of surface velocitiesThese are compar-able to values measured on this part of the glacier in pre-vious years (Harbor and others 1997) but somewhat lowerthan values reported from other glaciers (Raymond 1971Hooke and others1992)

As far as we know this is the first study to compare long-term englacial tiltmeter data with long-term inclinometrydata Tilt rates derived using both techniques correspondclosely (r2 ˆ094)

We have developed a technique for using englacial tilt-meter data to reconstruct internal velocity profiles overseasonal intervals This is useful as inclinometry cannotaccurately be used to derive deformation profiles over timeperiods less than about 1year on slow-moving glaciers(Copland and others 1997b) Our technique allows us todecompose surface velocities into internal deformation andbasal motion over time intervals ranging from days tomonthsThe technique assumes that englacial tiltmeter dataare representative of certain depth ranges and that theserepresentative depth ranges do not change between seasonsIt would be useful to test this assumption in future bydeploying more tiltmeters per borehole

Results indicate that at four locations covering an areasup1280 m along glacier and sup170 m across glacier patterns ofsurface velocity were similar between seasons even though

magnitudes were greater during spring events than duringsummer and autumnwinter magnitudes were lower stillHowever results show that patterns of internal deformationand basal motion differed between seasons In particularduring spring events high surface speeds were due to veryhigh basal motion in the upper and middle parts of thestudy area and high deformation rates in the lower partDuring the summer however internal deformation andbasal motion were approximately equal in the upper andlower parts of the study area but basal motion was muchsmaller than internal deformation in the middle part Thedifferences were even more marked in the autumnwinterwhen basal motion was greater than ice deformation in theupper and lower arrays but less than ice deformation in themiddle array The results can be interpreted in terms of sea-sonally migrating sticky and slippery spots Patterns of basaldrag appear to reflect changing water-pressure distribu-tions particularly the degree to which pressures approachor exceed ice overburden Our results compare favourablywith those from Storglacialaquo ren where variations in the pro-portions of internal deformation and basal motion weremeasured during three summers but contrast with thosefrom Worthington Glacier where no such variations weremeasured over a summer (Hooke and others 1992 Harperand others1998)

It is often assumed that basal motion drops to zero duringthe winter months An important conclusion of our work isthat while this might be true in certain regions (eg siteMB) it is not ubiquitous In fact basal motion may even in-crease during winter months compared with summer monthsin certain areas (eg site UB) This increase may be due toshort-term increases in water pressure lasting a few dayswhich may result from hydraulic adjustments as the subglacialdrainage system shrinks in response to declining discharges

ACKNOWLEDGEMENTS

The work was supported by UK Natural EnvironmentResearch Council grant GR311216 The paper was writtenwhile D Mair was in receipt of a Leverhulme Trust Fellow-ship We thank M J Sharp and H Blatter for stimulatingdiscussions during the formulation of this research and AHayes for help with the design and fabrication of the pres-sure transducers and englacial tiltmeters Fieldwork wascarried out with valiant help from K Arn G LeysingerR Middleton T Schuler D Swift and several others asso-ciated with the Arolla Glaciology ProjectWe thank GrandeDixence SAY Bams V Anzevui and P and B Bournissenfor logistical support in Switzerland The constructivereviews by G H Gudmundsson M Lulaquo thi N Span and AVieli helped to improve the original version of the paper

REFERENCES

Blake E G K C Clarke and M C Gerin 1992Tools for examining sub-glacial bed deformation J Glaciol 38(130) 388^396

Blatter H G K C Clarke andJ Colinge1998 Stress and velocity fields inglaciers Part II Sliding and basal stress distribution J Glaciol 44(148)457^466

Copland L J Harbor and M Sharp1997a Borehole video observationofenglacial and basal ice conditions in a temperate valley glacier AnnGlaciol 24 277^282

Copland L J Harbor M Minner and M Sharp1997bThe use of boreholeinclinometry in determining basal sliding and internal deformation atHaut Glacier drsquoArolla Switzerland Ann Glaciol 24 331^337

Gordon S M Sharp B Hubbard C Smart B Ketterling and I Willis


166

1998 Seasonalreorganizationof subglacial drainage inferred from mea-surements in boreholes Hydrol Processes 12105^133

Gordon S and 7 others 2001 Borehole drainage and its implications for theinvestigationof glacierhydrology experiences from HautGlacierdrsquoArollaHydrol Processes 15797^813

Gudmundsson G H1997 Basal-flow characteristics of a linear medium slid-ing frictionless over small bedrock undulations J Glaciol 43(143)71^79

Gudmundsson G H A Bauder M Lulaquo thi U H Fischer and M Funk1999 Estimating rates of basal motion and internal ice deformationfrom continuous tilt measurements Ann Glaciol 28 247^252

Harbor J M Sharp L Copland B Hubbard P Nienow and D Mair1997 The influence of subglacial drainage conditions on the velocitydistribution within a glacier cross section Geology 25(8)739^742

Harper JT N F Humphrey and WT Pfeffer 1998 Three-dimensionaldeformation measured in an Alaskanglacier Science 281(5381)1340^1342

Hooke R LeB P Holmlund and N R Iverson 1987 Extrusion flow demon-stratedby bore-hole deformation measurements overa riegel Storglacialaquo renSweden J Glaciol 33(113)72^78

HookeR LeB P Calla P Holmlund M Nilssonand A Stroeven1989A3 year record of seasonal variations in surface velocity Storglacialaquo renSweden J Glaciol 35(120) 235^247

Hooke R LeBV A Pohjola P Jansson andJ Kohler1992 Intra-seasonalchanges in deformation profiles revealed by borehole studies Stor-glacialaquo ren Sweden J Glaciol 38(130)348^358

Hubbard A H Blatter P Nienow D Mair and B Hubbard 1998 Com-parison of a three-dimensional model for glacier flow with field datafrom Haut Glacier drsquoArolla Switzerland J Glaciol 44(147) 368^378

Hubbard B 2002 Direct measurement of basal motion at a hard-beddedtemperate glacier Glacier de Transfleuron Switzerland J Glaciol48(160)1^8

Hubbard B P M J Sharp I C Willis M K Nielsen and C C Smart1995 Borehole water-level variations and the structure of the subglacialhydrological system of Haut Glacier drsquoArolla Valais Switzerland J

Glaciol 41(139) 572^583Hubbard B A Binley L Slater R Middleton and B Kulessa1998 Inter-

borehole electrical resistivity imaging of englacial drainage J Glaciol44(147) 429^434

Iken A and M Truffer 1997 The relationship between subglacial waterpressure and velocity of Findelengletscher Switzerland during itsadvance and retreat J Glaciol 43(144) 328^338

Kulessa B and B Hubbard 1997 Interpretation of borehole impulse testsat Haut Glacier drsquoArolla Switzerland Ann Glaciol 24 397^402

Lamb HR and8 others1995The compositionof subglacialmeltwater sampledfrom boreholes at the Haut Glacier drsquoArolla Switzerland InternationalAssociation of Hydrological Sciences Publication 228 (Symposium at Boulder1995oumlBiogeochemistry of SeasonallySnow-Covered Catchments)395^403

Mair D P Nienow IWillis and M Sharp2001 Spatial patterns of glaciermotion during a high-velocity event Haut Glacier drsquoArolla Switzer-land J Glaciol 47(156) 9^20

Nienow P M Sharp and IWillis1998 Seasonal changes in the morphologyof the subglacial drainage system Haut Glacier drsquoArolla SwitzerlandEarth Surf Processes Landforms 23(9) 825^843

Porter P R and T Murray 2001 Mechanical and hydraulic properties oftill beneath Bakaninbreen Svalbard J Glaciol 47(157)167^175

Raymond C F 1971 Flow in a transverse section of Athabasca GlacierAlberta Canada J Glaciol 10(58) 55^84

Sharp M and 6 others 1993 Geometry bed topography and drainagesystem structure of the Haut Glacier drsquoArolla Switzerland Earth SurfProcesses Landforms 18(6) 557^571

Tranter M H R Lamb M J Sharp G H Brown B P Hubbard and I CWillis 2002Geochemicalweatheringat the bedof Haut GlacierdrsquoArollaSwitzerland ouml a new model Hydrol Processes 16 959^993

Van derVeen C J and I MWhillans1993 Location of mechanical controlson Columbia Glacier Alaska USA prior to its rapid retreat Arct AlpRes 25(2) 99^105

167


aargletscher Switzerland over a 2 year period the highsensitivity of tiltmeters and their ease of manufacture meansthat the technique has the potential to work over large areasand over much shorter time intervals

The overall aim of this paper is to use surface motionmeasurements englacial tiltmeter records and repeat inclin-ometry data to determine patterns of surface internal andbasal motion across the tongue of a temperate valley glacierover a variety of temporal scales ranging from days tomonths We interpret the findings with reference to contem-poraneous measurements of subglacial water pressures andin terms of what we already know about the hydrology ofthe glacier

METHODS

Field site

Haut Glacier drsquoArolla is a sup163 km2 sup140 km long temper-ate valley glacier in Valais Switzerland (45sup358rsquo N 7sup332rsquo E)ranging in elevation from sup12560 to 3500 m asl (Fig 1)During each of the summers of 1992^99 an array of bore-holes was drilled sup115 km from the glacier snout and east ofthe glacier centre line (Fig 1) On average sup125 boreholeswere drilled each year ranging in depth from sup130 m nearthe margin to 4130 m on the centre line and these havebeen used to study the englacial and subglacial hydrology(B Hubbard and others 1995 1998 Copland and others1997a Kulessa and Hubbard1997 Gordon and others19982001) and geochemistry (Lamb and others1995Tranter andothers 2002) of this part of the glacier Boreholes drilled in1995 and 1996 were also used to calculate annual patternsof surface internal and basal motion in a cross-section of

the glacier from the centre line to the eastern margin usingrepeat inclinometry techniques (Harbor and others 1997)Analysis of several boreholes drilled in this area in 1998and1999 forms the basis of the work presented in this paperThe boreholes were drilled with reference to a subglacialdrainage axis that ran down the eastern half of the glacierduring the early to mid-1990s (Sharp and others 1993Hubbard and others 1995) The axis location is controlledlargely by the glacier geometry but also by the position ofmoulins Changes in glacier geometry and moulin positionmay have altered the precise location of this axis since it wasfirst identified but for the purposes of this paper we refer tothe axis as it existed in the early to mid-1990sThe axis trans-ports surface-derived waters andevolves from a distributed toa channelized morphologyduring the early summer (Gordonand others1998 Nienow and others1998) In late summer itconsists of a channel bounded laterally by subglacial sedi-ments During the day water and water-pressure waves areforced out of the channel into the surrounding sedimentsand possibly a thin film between the glacier sole and thesediments and at night water and water-pressure waves aredrivenback to the channel (Hubbard and others1995) Pres-sure fluctuations in the channel affect areas of the bed up tosup170 m either side (Hubbard and others1995)

Inclinometry

In late May and earlyJune1998 four boreholes were drilledto the bed at seven sites on the glacier tongue giving 28boreholes in total (Fig 1) Sites UA MA and LA werelocated along the centre of the subglacial drainage axisSites UB MB and LB were positioned sup170 m west of theaxis Site MC was located a further sup170 m to the west near

Fig 1 Haut Glacier drsquoArolla showing the location of the seven borehole sitesThe sites are labelled according to their position alongthe glacier (U upper M middle L lower) and across glacier (A over the axis C near the centre line B between the axis andthe centre line) Contours are shown for the glacier surface (solid) and bedrock (dashed)


158






Surface motion



Repeat inclinometry





m


159










160


161






162











163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167







Surface motion



Repeat inclinometry





m


159










160


161






162











163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167










160


161






162











163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167



161






162











163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167






162











163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167












163
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167
















sup3 a^1 sup3 a^1





MB 100^95 95^85 85^70 70^50 50^0MC 100^70 70^50 50^0LB 100^90 90^30 30^0UB 100^90 90^50 50^0



164









165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167










165



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167



CONCLUSIONS








ACKNOWLEDGEMENTS


REFERENCES







166

























167


























167


Date post:	24-Apr-2023
Category:	Documents
Upload:	aber
View:	0 times
Download:	0 times

Seasonal variations in ice deformation and basal motion across the tongue of Haut Glacier d'Arolla,...

Documents